KGS Home Geohydrology Home Reports

Kansas Geological Survey, Open-file Report 2002-60


Stream-aquifer interaction investigations on the Solomon River: Construction, geochemical sampling, and slug testing of groundwater observation wells in Rooks County, Kansas and geologic logging of existing wells in Rooks, Smith and Osborne County, Kansas

by
James J. Butler, Jr., Donald O. Whittemore, John M. Healey, and Marcia K. Schulmeister


KGS Open-file Report 2002-60
December 2002

Executive Summary

Eleven direct-push electrical-conductivity logs were performed by the Kansas Geological Survey (KGS) as part of an effort to better understand the lithology of the unconsolidated alluvial sediments adjacent to the North and South Forks of the Solomon River in Rooks, Smith, and Osborne counties in north-central Kansas. Three observation wells were constructed by the KGS near the United States Geological Survey stream-gaging station upstream of Webster Reservoir in Rooks County to clarify the relationship between the Solomon River and the adjacent alluvial aquifer. The direct-push logs were used to select appropriate intervals for well screens and annular seals at those wells. Following well construction and development, water samples were obtained from each well and a series of slug tests were performed.

The electrical-conductivity logs were interpreted assuming that the observed variations in electrical conductivity were a product of variations in lithology. The logs revealed that at most sites there was a considerable thickness of interbedded clays and silts overlying a sand interval immediately above the bedrock. At some locations, sands were also interbedded with the overlying clays and silts. The depth to bedrock varied significantly between logs. Although the electrical-conductivity profiles from a pair of logs separated by less than two feet were in excellent agreement, significant variations were seen between profiles separated by greater distances.

Ground waters from the two observation wells constructed north of the South Fork of the Solomon River are fresh, although the upper part of the alluvial aquifer (well RO18) contains a higher total dissolved solids (TDS) concentration than in the lower part of the aquifer (well RO19). The higher TDS content of the shallower water could be the result of evapotranspiration consumption of irrigation water, leaving residual salts in a smaller volume of water. Alternatively, seasonal pumping of the aquifer north of the South Fork of the Solomon River might have induced infiltration of river water into the alluvial aquifer, especially the shallow portion, that then mixed with existing ground water. A combination of both stream-aquifer interaction and irrigation return recharge could also be a possible explanation for the major dissolved constituent contents. The nitrate concentration in the upper portion of the aquifer is substantially greater than in the lower portion, although still well below the drinking water standard. The higher nitrate content in the shallow part of the aquifer, which is most probably from agricultural sources, supports the mechanism of infiltration of water from the surface. Ground water in the well constructed on the south side of the South Fork of the Solomon River (well RO20) is saline but contains very low nitrate content. The chemistry is not consistent with river water. Phreatophyte evapotranspiration of shallow ground water that derived most of its dissolved constituents from bedrock and upland soil sources is the most logical control on the ground-water chemistry at well RO20. The TDS, sulfate, and chloride concentrations of the South Fork Solomon River above Webster Reservoir have all increased substantially from the period 1963-1975 to 1990-1998.

The slug tests were performed and interpreted using methods developed or refined at the KGS. All of the slug tests appeared to have been influenced by mechanisms associated with slug tests in highly permeable aquifers, and were therefore analyzed using models that incorporated those mechanisms. The hydraulic-conductivity estimates obtained from the slug tests are characteristic of the coarse sands and gravels that would be expected in alluvial aquifers. The highest value for hydraulic conductivity (241 feet/day) was obtained at RO20 south of the Solomon River, and was over twice the values for hydraulic conductivity obtained at the two wells north of the river (RO18 (84 feet/day) and RO19 (118 feet/day)).

Introduction

Eleven direct-push electrical conductivity logs were performed by the Kansas Geological Survey (KGS) at sites in Rooks, Smith, and Osborne counties in north-central Kansas. The electrical-conductivity logs were used to better define the lithology of the unconsolidated alluvial sediments adjacent to the Solomon River and to design three observation wells that were constructed by the KGS in the vicinity of the United States Geological Survey (USGS) stream-gaging station immediately upstream of Webster Reservoir in Rooks County. Following well construction and development, water samples were collected from each well and a series of slug tests were performed. Funding for the project was provided by the State Water Plan through the Subbasin Water Resource Management Program of the Division of Water Resources of the Kansas Department of Agriculture. This project is part of an effort to better understand the lithology of the alluvial aquifers adjacent to the North and South Forks of the Solomon River, and the relationship between the South Fork of the Solomon River and the alluvial aquifer upstream of Webster Reservoir. Jim Butler of the Geohydrology Section of the KGS served as the principal investigator for the project.

Report Overview

The following report will be divided into three main sections: Preliminary Site Investigation, Well Construction and Development, and Water Chemistry and Aquifer Hydraulics Investigations.

The first section describes the preliminary site investigation activities (direct-push electrical conductivity logging) that were used to assess subsurface geology and to determine the screened intervals for the observation wells. The second section describes the details of the well installation and the methods that were used to develop the installed wells. The third section includes a description of the field procedures used for acquiring water samples for geochemical analyses and for performing slug tests to obtain estimates of hydraulic conductivity. The description of the field procedures is followed by an interpretative discussion.

Preliminary Site Investigation (April 24-25, 2002; July 16, 2002)

Electrical Conductivity Surveys

Methodology Overview

The preliminary site investigation involved use of a direct-push technique commonly referred to as electrical-conductivity profiling (Christy et al., 1994; Butler et al., 1999). This technique is effective in defining subsurface lithology in unconsolidated alluvial sediments. The electrical-conductivity (EC) probe used in this work (Geoprobe SC 400) was designed such that the investigator can select different configurations to adjust the lateral extent of investigation. The configuration that gave the greatest lateral penetration (Wenner array) was used here.

The EC probe was advanced into the subsurface with a track-mounted Geoprobe 66DT unit using hydraulic pressure and a percussion hammer, while the electrical conductivity of the subsurface was measured at intervals of 0.05 ft. Unless noted otherwise, the electrical conductivity of the unconsolidated sediments was assumed to be primarily a function of grain size. Thus, low values of electrical conductivity were assumed to indicate sand and gravel, intermediate values were assumed to indicate silt, and high values were assumed to indicate clay. Although no cores were taken at these sites to confirm the assumed relationships, previous work at a KGS research site has shown the general viability of these relationships (Butler et al., 1999).

The EC probe was advanced until either the tool could be driven no further, often indicative of contact with bedrock or other well-indurated material, or the depth was below the interval of interest. The information obtained by the EC logs was the only means of lithologic interpretation used in this work because the well installation procedure did not produce drill cuttings for visual inspection and description, and no cores were taken.

The KGS was contracted to complete eight EC logs. However, additional logs were completed near Osborne and near the USGS gaging station above Webster Reservoir to enhance the understanding of the near-surface geology in those areas. The results from the EC log at each site will be discussed separately, individual logs will be plotted in the format of electrical conductivity in milliSiemens/meter (mS/m) versus depth in feet. Plots are shown with respect to land surface and are not corrected for elevation differences. To make this correction, the exact elevation for each probe hole, which was unavailable at the time of the report, would be needed.

Osborne County Sites

Four EC profiles were obtained near the city of Osborne in Osborne County.

Osborne North Sites (OB1 and OB2)

The Osborne North 1 site (OB1--NE, NE, NE Sec. 18, T. 7 S., R. 12 W.) is located on the north side of the city of Osborne approximately 1.7 miles north of the active channel of the South Fork of the Solomon River. The probe hole was located just southwest of the intersection of highways U.S. 24 and U.S. 281 in the lawn of a motel, 7.8 feet due west of an existing well. The Osborne North 2 site (OB2--SW, NW, NW Sec. 17, T. 7 S., R. 12 W.) is located approximately 0.2 miles south of the intersection of highways U.S. 24 and U.S. 281 on the east side of U.S. 281. The probe hole was located just north of the intersection of U.S. 281 and Mercury Avenue, 22 feet north of an existing well.

The two Osborne North EC profiles (OB1 and OB2) show a sharp drop in electrical conductivity at about 50 ft. Above this drop, the EC profiles are characterized by high to intermediate values, indicating an interval of clay with infrequent silt layers. Below this depth, an 11-12 ft zone of lower EC values is observed, indicating the presence of sand and gravel with varying amounts of silt. The position of this coarser unit, as well as that of the inferred top of bedrock (increase in EC below the sand and gravel), appears to be about 5 ft higher in OB2 than in OB1. These differences in elevation are most likely a result of differences in the land-surface elevation of the log locations. Direct comparison of the logs should not be made until surface elevations have been determined for the log locations.

Both profiles show drop around 50 feet; OB1 has values higher than OB2 in much of profile above 50 ft.

Osborne South Sites (OB3 and OB4)

The Osborne South 1 site (OB3--SE, SE, NE Sec. 19, T. 7 S., R. 12 W.) is located on the south side of the city of Osborne approximately 1.5 miles south of OB1 on the east side of U.S. 281, just south of an unnamed intermittent creek that feeds into the Solomon River to the southeast. The probe hole was located 3.1 feet due west of an existing well that is just northeast of a car wash.

The Osborne South 2 site (OB4--SE, SE, SE Sec. 19, T. 7 S., R. 12 W.) is located 1.9 miles south of OB1 on the east side of U.S. 281. The probe hole was located on the grounds of a veterinary clinic, 10.4 feet southwest of an existing well and 3.4 feet northeast of the closest leg of a nearby windmill. This was the only EC log in the vicinity of Osborne obtained south of the Solomon River.

The EC profile for Osborne South 1 (OB3) indicates a soil zone at the surface followed by an interbedded sequence of sands and silt to a depth of 6 feet. A clay-rich zone extends from about 6 to 15 feet below land surface, followed by a sequence of sands and silts to a unit of high electrical conductivity at approximately 27.8 feet below land surface. This unit of high electrical conductivity was thought to be shale, but further work would be needed to verify that assumption.

Profile shows drop around 20 ft and higher values at 27-30 ft.

The EC profile for Osborne South 2 (OB4) shows interbedded sands and silts at shallow depths, followed by a predominantly clay sequence from approximately 6.8 to 28.7 feet in depth. Intermediate EC values in the interval from 28.7-50.5 feet suggest a predominance of silt with occasional clay and sand lenses. A sand-rich interval appears between the clay bed (50.5-52.9 feet) and bedrock (57.6 feet). The presence of this sandy interval immediately above the bedrock, as well as the depth to bedrock, is consistent with the OB1 and OB2 logs.

Profile shows drop around 30 ft, seems to stay low except for small jump at 50 ft.

Smith County Site

A single EC log (SM1--SE, SE, SE Sec. 25, T. 5 S., R. 13 W.) was obtained in Smith County approximately 1.5 miles northwest of the city of Portis. The probe hole was 3.9 feet due north of an existing well and 3.8 feet northeast of the USGS benchmark that is just northwest of the road intersection.

The EC profile for the Smith County site shows interbedded sands and silts from the land surface until a depth of about 10 feet, followed by a clay and silt sequence to a depth of approximately 42 feet. A sand and gravel interval appears between the bottom of the clay and silt sequence and the inferred top of bedrock at 56 feet. The geologic configuration of a thick clay and silt sequence overlying a coarser interval immediately above bedrock is similar to that observed in the two Osborne North logs (OB1 and OB2).

Profile shows gradual increase from 0 to 40 ft; drops and stays low until a jump up at around 60 ft.

Rooks County Sites

Six EC profiles were obtained in the vicinity of the USGS stream-gaging station on the South Fork of the Solomon River above Webster Reservoir. Five profiles were obtained in the initial preliminary investigation and the sixth was obtained immediately prior to drilling of the observation wells. All profiles were obtained on the west side of the asphalt road running north from Damar to U.S. 24 (henceforth, Damar Road).

North of Observation Wells (RO1, RO2, and RO2B)

Three EC profiles were obtained along Damar Road to the north of the observation wells. The northernmost log (RO1--NE, NE, NE Sec. 7, T. 8 S., R. 20 W.) is 26.3 feet due west of the white shoulder line on the west side of Damar Road approximately 0.2 miles north of observation wells RO18 and RO19. Two additional logs (RO2 and RO2B--SE, NE, NE Sec. 7, T. 8 S., R. 20 W.) were obtained 1.8 feet apart just to the west of the north-south fence line on the west side of Damar Road approximately 0.1 miles north of wells RO18 and RO19.

The EC profile at RO1 indicates sandy sediments to a depth of about 6 feet and silts and clay from 6 to 14 feet. A sand interval of approximately 8 feet in thickness is directly above the bedrock surface, which was encountered at a depth of 22 feet.

Profile is low except for peak from 6 to 14 ft, sharp jump at 21.

The two EC logs obtained at the RO2 location (RO2 and RO2B) are in good agreement. The unconsolidated section is dominated by sand with three distinct clay lenses at depths of about 7, 13, and 19 feet below land surface. The similarity of the closely spaced logs demonstrates the ability of the EC probe to reproduce the details of the subsurface lithology. Although the depth to bedrock encountered by the RO2 logs is similar to that observed in RO1, the differences in the EC logs at the two sites demonstrates that geologic variability occurs over relatively short distances north of the Solomon River.

Two profiles are very similar; RO2B has a much higher spike at 18-19 ft.

Site of Observation Wells RO18 and RO19 (Logs RO3 and RO5)

Two EC profiles were obtained near the location of wells RO18 and RO19. The first probe hole (RO3--NE, SE, NE Sec. 7, T. 8 S., R. 20 W.) is located south of an east-west trending agricultural access road that passes between two windmills, 5.2 feet due north of the northern legs of the southern-most windmill. The second probe hole (RO5--NE, SE, NE Sec. 7, T. 8 S., R. 20 W.) is located approximately 220 feet to the east of the first probe hole at the present site of well RO18. The second log was obtained because the site of the first log was deemed too close to the windmill for the construction of monitoring wells used for water quality analyses.

Although the two EC profiles were obtained only a few hundred feet apart, significant lithologic changes occurred between the two profile locations. In profile RO3, an upper six feet of sand is followed by interbedded silts and sands that extend to approximately 18 feet in depth. The EC profile indicates a clay-rich zone from 18 to 34 feet followed by a predominantly silt interval from 34 to 53 feet in depth. A sand interval of approximately 5 feet in thickness overlies the inferred top of bedrock (58 feet below land surface). In profile RO5, an upper six feet of sand is followed by a silt and sand interval to approximately 23 feet in depth. The EC log indicates a clay and silt zone from 23-32 feet in depth followed by thin sand (32-36.5 feet in depth) and clay (36.5-40 feet in depth) intervals. The lower portion of the unconsolidated sequence at RO5 consists of a sand zone from 40-47 feet, a clay and silt interval from 47-52 feet, and a sand interval of approximately seven feet in thickness overlying the bedrock surface (59 feet below land surface). Although the logs differ significantly, they both indicate a shallow interval of interbedded silts and sands and a sand interval immediately above the bedrock. These two intervals were selected as the screened intervals for observation wells RO19 and RO18, respectively.

Two profiles similar, but RO5 seems to read lower in most cases than RO3; RO5 has a spike at 37 ft not seen as well in RO3.

Site of Observation Well RO20 (Log RO4)

One EC log was obtained at the location of well RO20 south of the Solomon River on the west side of the Damar Road (RO4--SE, SE, NE Sec. 7, T. 8 S., R. 20 W.). EC log RO4 was obtained 65.7 feet due west of the south edge of the bridge railing and 44.3 feet northeast of an existing well. The probe hole was located just west of a north-south fence line.

The EC profile at RO4 indicates topsoil to a depth of about 2.5 feet followed by an interval of interbedded clays, silts, and sands to 16 feet. A sand interval of approximately 7 feet in thickness is directly above the bedrock surface, which was encountered at a depth of 23 feet. Observation well RO20 was screened in this lower sand. Note that the electrical conductivity of this sand interval is higher than that observed at the other logging locations as a result of the high specific conductance of the ground water (see Table B in section entitled Chemical Quality of Observation Well Waters).

Sharp spike at 5 ft, others at 7, 10; lower after 15 ft.

Well Construction and Development (July 16-17, 2002)

Well Construction

All of the observation wells were completed in accordance with regulations established by the Kansas Department of Health and Environment. Well depths and screened intervals were chosen based on the EC logs and discussions with DWR personnel. Copies of the well completion forms (WWC-5) are included at the end of this report. The geologic interpretations given on the WWC-5 forms are based upon the electrical-conductivity profiles obtained at each site.

All wells were constructed in the same manner with the KGS Geoprobe unit (66DT). The unit was used to drive 3.5-inch (outer diameter) flush-joint steel casing (type NW) with inner rods and pilot bit to the desired depth. Upon reaching that depth, the inner rods were extracted while the 3.5-inch NW casing was flooded with potable water to prevent sand heave. The total depth inside the NW casing was measured to ensure correct well screen placement.

Each well was constructed of 2-inch (Sch. 40 PVC) flush-joint casing with 5 feet (nominal) of 2-inch (Sch. 40 PVC) 20-slot screen with a well point at the lower end. The casing string was assembled as it was lowered through the center of the NW pipe. After the PVC casing string, with the screen and well point at the lower end, was lowered to the bottom, the outer NW casing was retracted. The unconsolidated formation collapsed back against the PVC casing and screen as the NW pipe was removed. Retraction of the steel pipe continued until the bottom of the pipe string was adjacent to a zone where an annular seal was necessary. Bentonite slurry was then tremied into the annular space between the inner diameter of the NW pipe and the outer diameter of the PVC casing. After the NW pipe was removed from the hole, bentonite chips were poured into the borehole to fill the remaining annular space to the land surface. Placement of a 4-inch steel well protector with a lock over the PVC casing completed the well. The following diagrams show the well-construction details, the electrical-conductivity profiles, and a brief geologic interpretation for each well site. A table summarizing well details is provided following the diagrams.

Log, interpretation and construction details.

Log, interpretation and construction details.

Table of Observation Well Details
Location
County/Well Name
Legal Description Total Depth
(post-development
from grade)
Approximate
Screen
Interval
(from grade)
Approximate
Interval for
Annular Seals
(from grade)
Water Level During the
3rd week of July 2002
(post-development wrt TOC)
Rooks Co.,
RO18
Township 8 South,
Range 20 West,
Section 7
NE, SE, NE
58.29 53-58 0-16 16.33
Rooks Co.,
RO19
Township 8 South,
Range 20 West,
Section 7
NE, SE, NE
23.50 18-23 0-14.5 16.03
Rooks Co.,
RO20
Township 8 South,
Range 20 West,
Section 7
SE, SE, NE
22.43 17-22 0-10 10.58
All measurements are given in feet.

All wells are constructed of two-inch Schedule 40 flush-joint casing with five
feet (nominal) of two-inch 20-slot screen and a well point. Top of casing
(TOC) is 1.8, 1.6, and 2.0 feet above land surface (grade) for wells RO18,
RO19, and RO20, respectively.

Well Development

The completed wells were developed using a centrifugal suction pump capable of flow rates of over 10 gallons per minute (gpm). While the pump was operating, the wells were mechanically surged by rapidly raising and lowering the pump intake line, which had a check valve at its lower end. The pumping and surging continued until groundwater turbidity was judged minimal. At that point, the intake line was removed from the well and a small-diameter Grundfos submersible pump was installed for sampling.

Water Chemistry and Aquifer Hydraulics Investigations

Chemical Quality of Observation Well Waters

The KGS collected water samples from the three observation wells on July 17, 2002. All three wells were developed prior to sampling. In all cases, the water after development was clear. The ground waters were sampled using a Grundfos pump operating at a rate of about 8 gpm. Wells RO18, RO19, and RO20 were pumped for 45, 71, and 50 minutes before sampling, respectively. There was no significant change in the temperature and specific conductance after the first several minutes of pumping wells RO18 and RO20, and in the temperature during pumping well RO19. However, the field specific conductance of water from well RO19 slowly declined from 1,184 μS/cm (equivalent to μmho/cm) at 5 minutes of pumping to 1,148 μS/cm just before sampling. Although the change was only about 3%, the slow, steady decline in conductance suggests that the pumping might reflect small differences in the dissolved solids concentrations of ground waters at different depths in the zone from which the water was pumped.

The samples from the wells were filtered through 0.45-μm membrane filter cartridges in the field. A portion of the field-filtered samples was acidified with hydrochloric acid for use in cation and nitrate determination. All samples were immediately placed in a cooler with ice and transferred to a refrigerator after arrival at the laboratory. The concentrations of silica, cations, and boron were determined using inductively-coupled plasma spectrophotometry. Alkalinity was measured by automated titrimetry and converted to bicarbonate content. Sulfate, chloride, and nitrate concentrations were determined using colorimetric or ultraviolet spectrophotometry on automated flow-injection or segmented-flow instruments. Fluoride content was measured with a specific-ion electrode. The collection information and chemical properties of the water samples are listed in Table A; the dissolved constituent concentrations and selected ratios are in Table B.

Table A. Location Information and Chemical Properties for Water Samples Collected from the Observation Wells. Sp.C. refers to specific conductance (μS/cm is equivalent to μmho/cm).

Site name Co. T.R.S. Location KGS
lab no.
Project
number
Sample
date
Sample
time
Field
Sp.C.
μS/cm
Lab
Sp.C.
μS/cm
Lab
pH
Well RO-18 RO 08S-20W-07A 020219 RO-001 7/17/02 15:42 791 790 7.50
Well RO-19 RO 08S-20W-07A 020220 RO-002 7/17/02 14:46 1148 1145 7.20
Well RO-20 RO 08S-20W-07ADDD 020221 RO-003 7/17/02 17:06 2930 2870 7.00

Table B. Dissolved Constituent Concentrations and Ratios for the Observation Well Waters. TDS and SpC refer to total dissolved solids and specific conductance, respectively. The TDS was computed as the sum of dissolved constituents.

Site name SiO2
mg/L
Ca
mg/L
Mg
mg/L
Na
mg/L
K
mg/L
Sr
mg/L
HCO3
mg/L
SO4
mg/L
Cl
mg/L
F
mg/L
NO3-N
mg/L
B
mg/L
TDS
mg/L
TDS/
SpC
Na/Cl
equiv.
ratio
(Ca+Mg)/Na
equv.
ratio
SO4/Cl
equiv.
ratio
Well RO-18 57.8 102 16.8 29.9 11.6 0.94 265 134 35.9 0.53 0.1 <0.02 520 0.658 1.284 4.98 2.75
Well RO-19 37.2 153 25.5 45.1 11.4 1.14 268 232 79.8 0.55 4.2 <0.02 718 0.627 0.871 4.96 2.15
Well RO-20 45.9 551 46.1 111 10.3 4.63 455 1091 169 0.88 0.4 0.20 2254 0.785 1.013 6.48 4.76

Wells RO18 and RO19 are at the same location but are screened over different intervals in the alluvial aquifer. The total dissolved solids (TDS) content of samples from both of these wells indicates that the ground water is fresh (less than 1,000 mg/L) at this location north of the South Fork of the Solomon River. However, the TDS concentration of the shallow well RO19 (24 ft deep) is substantially greater than that of the well at the bottom of the alluvial aquifer (58 ft deep). The TDS content of the ground water from the shallow well south of the river (RO20) is much greater than the TDS of the ground waters at both the shallow and deep parts of the aquifer north of the river (wells RO-18 and RO-19). Although the concentrations of all the major dissolved cations and anions (calcium, magnesium, sodium, bicarbonate, sulfate, and chloride) in the observation well waters increase with the TDS content, the concentrations of calcium and sulfate increase much more than those of the other constituents. This is reflected in the change in the chemical water type from calcium-bicarbonate for the ground water from the deep part of the aquifer north of the river (RO18) to calcium-sulfate, bicarbonate for the shallow aquifer north of the river (RO19) to calcium-sulfate for the aquifer south of the river (RO20).

The greater concentrations of the major dissolved constituents and of nitrate in the shallow ground water than in the deep ground water north of the river suggests that the shallow ground water could represent the impact of irrigation water recharge. The lower TDS in the deep portion of the alluvial aquifer indicates that the bedrock does not appear to be a source of high dissolved solids contents. Waters used for irrigation are pumped from the aquifer in the river valley in the general area of the study. The dissolved solids content of irrigation water spread at the surface would be concentrated by evaporation and plant transpiration as water is consumed and dissolved salts are left in the remaining water. The greater potential evapotranspiration than precipitation in this area of Kansas means that the soil water would not be diluted enough by precipitation recharge to completely counteract the evapotranspiration concentration of salts in the soil solution. The higher nitrate concentration is consistent with a surface agricultural source for the area. The sulfate/chloride ratio of the shallow ground water at well RO19 is slightly less than that for water from well RO18. This is consistent with evapotranspiration of ground water pumped from the lower part of the aquifer followed by retention of some sulfate in the soil relative to chloride. Sulfate salts are much less soluble than chloride salts and could be precipitated during dry periods in soils; chloride salts in dry soils would also be expected to dissolve faster than sulfate salts during leaching by irrigation water and rainfall.

The ground water from well RO20 in the alluvial aquifer south of the river does not have a chemistry that is consistent with evapotranspiration concentration of dissolved salts in the proportions present in the shallow or deep ground water north of the river. The primary indicator of the difference is the substantially greater sulfate/chloride ratio for water from well RO20 than that for water from wells RO18 and RO19. If the high TDS in the ground water in the alluvial aquifer south of the river were derived from evapotranspiration concentration of dissolved salts in ground water similar to that north of the river, the sulfate/chloride ratio would be expected to either be relatively similar or somewhat lower than the ratios for ground water north of the river based on the reasons described at the end of the previous paragraph.

The bedrock underlying the alluvial sediments of the study area is inferred to be the Niobrara Chalk. The water from well RO19 screened in the alluvial aquifer just above the bedrock north of the river has a relatively low TDS content, suggesting that ground water in the bedrock does not have a high TDS concentration if it is assumed that some flow of water from the bedrock enters the alluvial aquifer at this location. Prescott (1955) reported analyses for four wells in eastern Graham County and Schmidt and Marsi (1961) reported analyses for two wells in western Rooks County in the Niobrara Chalk; the well water did not appear to be contaminated by surface waste (NO3-N was less than 3 mg/L). The sulfate and chloride concentration ranges for the six Niobrara wells were 46-335 mg/L and 28-173 mg/L, respectively. The higher concentrations for Niobrara ground water were for wells between 300 and 400 ft. Niobrara wells less than 200 ft deep contained sulfate contents less than 170 mg/L. KGS publications on the Niobrara do not indicate that there is any gypsum in the formation. Thus, it does not appear that a direct bedrock source could be an explanation for the high sulfate and TDS contents of the water from well RO20.

Well RO20 is much closer to the river than wells RO18 and RO19. Therefore, water chemistry data for the South Fork of the Solomon River at the location were compared to that from well RO20 to consider whether the high TDS source is river water. As indicated below in a discussion of the river-water chemistry, the TDS content and sulfate/chloride ratio of past river water at the site appear to be too low to be the direct source of the high TDS concentration. The much lower sulfate/chloride ratio of the river water than in well RO20 water suggests that evapotranspiration concentration of dissolved salts in ground water dominated by river-water chemistry is not the primary mechanism generating the high TDS content.

The study area is located in far western Rooks County only a mile east of the Graham County line. Chemical data for ground water in the alluvial aquifer along the South Fork of the Solomon River in T8 S, R21 W in far eastern Graham County are reported in Prescott (1955) and on the web pages of the USGS. There are two wells to the south of the river for which data were found. One of the wells (NW, NE, NE Section 17, T. 8 S., R. 21 W., a supply well in the town of Bogue) was subsequently sampled nine times up through 1989 and showed substantial change in chemistry. The earliest analysis of the Bogue well water (1953) and the analysis in 1952 for the other well (SW, SW, SE Section 8, T. 8 S., R. 21 W., domestic use) were used as the most representative of ground water least affected by activities associated with the town and irrigation, including induced recharge from the river caused by pumping of ground water. In addition, Schmidt and Marsi (1961) reported an analysis of water from a 2.5 feet deep hole in the alluvium in the same 40-acre tract as for well RO20.

The sulfate/chloride ratios for these three historical analyses were examined to determine whether evapotranspiration concentration of dissolved constituents in alluvial ground water south of the river with a similar chemistry could produce the observed chemistry at RO20. Well RO20 is in an area of riparian vegetation in the river valley. The sulfate/chloride ratio (based on equivalent concentrations) ranged from 3.7 to 5.5 for the three prior analyses compared to 6.48 for the RO20 water. Sulfate/chloride equivalent ratios for six ground waters from the Niobrara in western Rooks and eastern Graham counties (see above) ranged from 0.84 to 3.9. The ratios for the alluvial ground water south of the river and the highest ratio for the Niobrara ground water are in the same range as, or at least much closer to, that for the RO20 water (SO4/Cl equivalent ratio of 4.76) than the 2.15-2.75 range for wells RO18 and RO19. Therefore, the hypothesis of evapotranspiration concentration of dissolved solids in the alluvial ground water on the south side of the river by phreatophytes generally fits the chemical data and the location. The main sources of the sulfate, chloride, and other dissolved constituents in the ground water before concentration would most likely be flow through the Niobrara bedrock to the alluvium, as well as flow through soil and loess deposits upslope of the alluvial valley. The conceptual model involves flow from the alluvium south of the river towards the river most of the time, although some infiltration of river water would have occurred from flood stages in the past.

The uppermost 4 ft of the soil and sediment at well RO20 site is sandy. Some clay lenses are interbedded with silt and sand in the depth interval 4-16 ft. The upper sandy zone could allow rainfall runoff from the upland to readily infiltrate into the subsurface. The clays would retard the downward movement and possibly cause perching of some of the recharge. The shallower roots of phreatophytes could consume this water and leave most of the dissolved salts in the soil and sediment. Later recharge could dissolve the accumulated salts; some of the recharge could infiltrate deeper into the profile. This process, along with the consumption of deeper ground water by phreatophyte roots, could slowly increase the dissolved solids of the ground water. The process would be expected to occur all along the riparian zone. Thus, ground-water flow down the river valley under the phreatophyte area could be expected to bring water with some prior increase in dissolved solids to the location of well RO20.

There is a stock well near well RO20 and a cattle pen is in the area. Although waste from cattle could possibly contribute to the salinity of the ground water, the amount is expected to have an insignificant impact on the salinity because the nitrate content of the RO20 well water is so low.

Chemical Quality of the South Fork of the Solomon River and Stream-Aquifer Interactions

The South Fork of the Solomon River was not flowing at the time of ground-water sampling and did not have measurable flow up to the preparation date of this report. Thus, a river sample could not be obtained. Additional work was conducted for the project to assemble and analyze a dataset of existing water-quality data for the river. Existing data for analyses of river-water quality and flow for the South Fork of the Solomon River above Webster Reservoir (for the river at the gaging station in the study area) were obtained from USGS and EPA databases. The EPA data are for the Kansas Department of Health and Environment (KDHE) stream-monitoring network in Kansas. There are 97 analyses for the period 1963-1975 (USGS data) and 42 analyses for 1990-1998 (KDHE data) that include major constituent concentrations. The TDS, sulfate, and chloride concentration ranges for the entire data set are 213-1110 mg/L, 19-430 mg/L, and 7-205 mg/L, respectively. The mean and median are 639 and 585 mg/L for TDS, 207 and 190 mg/L for sulfate, and 64 and 48 for chloride, respectively. The range, mean, and median for the sulfate/chloride equivalent ratio are 1.24-3.69, 2.54, and 2.58, respectively.

TDS looks lower overall for 1963-1975 period than 1990-1998 samples, for similar flows.

Sulfate looks lower overall for 1963-1975 period than 1990-1998 samples, for similar flows.

In general, the concentration of TDS, sulfate, and other major dissolved constituents decrease with increasing flow in the South Fork of the Solomon River above Webster Reservoir. However, there is a shift to higher contents for similar flows for the 1990-1998 data set in comparison with the 1963-1975 data set as indicated by the pattern of points and the best fit lines (power functions) in the TDS and sulfide plots. The median TDS, sulfate, and chloride concentrations were 552, 170, and 41 mg/L for 1963-1975, and 839, 292, and 102 mg/L for 1990-1998. As shown in the following plot, the sulfate/chloride ratio appears to decrease slightly with increasing river flow but the relationship to flow does not appear to be significantly different for the two time periods.

Ratio of sulfate to chloride does not seem to change with flow much; two time periods have very similar responses.

However, the range, mean, and median sulfate/chloride ratio were lower for 1990-1998 in comparison to 1963-1975.

On average, 1990-1998 data have lower ratio than 1963-1975, though the difference is not dramatic; both have low ratio points, later period has fewer high-ratio points.

As illustrated in the following figure, the distribution of river flow with time for the entire gaging record shows a slow decreasing trend to the early 1990's followed by generally higher flows similar to those in the late 1950's to late 1960's (zero values of flow are not shown on log axis, although they occur during most years of the flow record). There is no clear difference in the pattern of flow values (filled circles on following plot) for the dates of the water samples for the 1963-1975 and 1990-1998 periods, although the best-fit line does indicate a small decrease in flow with time.

Flow slightly decreases with time until early 1990s, when there is a jump in flow.

Slight decrease in flow over time; later period seems to have more very low values and similar numbers of medium values.

Slug Test Methodology

The program of slug tests was carried out following guidelines for the design, performance and analysis of slug tests developed at the KGS (Butler et al., 1996; Butler, 1998; Fetter, 2001). The pneumatic method (Butler, 1998) was used for test initiation at all wells. This method involves placing an airtight wellhead apparatus on top of the well and pressurizing the air column in the sealed well casing. A slug test is initiated by a very rapid depressurization of the air column using a release valve. At least seven tests were performed at each well. The initial water-level change (H0) was varied in a systematic manner between tests so that the dependence of test responses on H0 and the significance of dynamic skin effects (Butler, 1998) could be assessed. As shown in the figures, head dependence was found to be negligible at all wells. A small amount of dynamic skin effects were observed at RO20 indicating that additional well development continued during the program of slug tests. However, the impact on the hydraulic conductivity estimate at RO20 was considered to be insignificant.

Changes in water level were measured using a pressure transducer (In-Situ PXD-261 0-20 psig transducer) connected to a data logger (Campbell Scientific CR23X). Air pressure within the sealed casing was also monitored using a pressure transducer (In-Situ PXD-261 0-20 psig transducer). Casing pressurization was accomplished using a nitrogen gas cylinder.

Tests at all three wells were influenced by inertial mechanisms associated with slug tests in highly permeable aquifers (Butler, 1998). Test data were therefore analyzed using two theoretical models for slug tests in highly permeable aquifers that incorporated those mechanisms: 1) the high-K Bouwer and Rice model for slug tests with inertial effects in partially penetrating wells in unconfined aquifers (Springer and Gelhar, 1991; Butler and Garnett, 2000); and 2) the high-K Hvorslev model for slug tests with inertial effects in partially penetrating wells in confined aquifers (Butler and Garnett, 2000). All analyses were performed with AQTESOLV, an automated well-test analysis package (HydroSOLVE, 2001). The choice of confined or unconfined model was made on the basis of electrical-conductivity logs and water-level measurements. The effect of vertically adjacent clay and silt intervals was incorporated in both models.

Response of well RO18 to 7 tests.

Response of well RO19 to 9 tests.

Response of well RO20 to 10 tests.

Slug Test Results

The hydraulic conductivity (K) estimates obtained at all three wells are characteristic of interbedded coarse sands and gravels (Fetter, 2001), consistent with the expected composition of the alluvial aquifer. RO20 has a considerably higher K than the other two wells, indicating the presence of coarser materials in the vicinity of that well. Note that Butler (1998) emphasizes that the impact of incomplete well development on slug tests will be difficult to avoid and can lead to K estimates that are significantly lower than the actual hydraulic conductivity of the aquifer. However, considerable attention was paid to well development in this work, so the tabulated values for hydraulic conductivity should be reasonable estimates of the hydraulic conductivity of the aquifer in the vicinity of the test wells.

Table of Hydraulic Conductivity Estimates Obtained from Slug Tests
Well
Location
Test Date Test
Number
Theoretical Model Hydraulic
Conductivity
(feet/day)
RO18 7/17/02 1 High-K Hvorslev 84
RO19 7/17/02 3 High-K Bouwer and Rice 118
RO20 7/17/02 10 High-K Hvorslev 241

References

Butler, J.J., Jr., 1998, The Design, Performance, and Analysis of Slug Tests: Lewis Publishers.

Butler, J.J., Jr., and Garnett, E.J., 2000, Simple procedures for analysis of slug tests in formations of high hydraulic conductivity using spreadsheet and scientific graphics software: Kansas Geol. Surv., Open-File Rep. 2000-40. [available online]

Butler, J.J., Jr., McElwee, C.D., and Liu, W.Z., 1996, Improving the reliability of parameter estimates obtained from slug tests: Ground Water 34, no. 3, p. 480-490.

Butler, J.J., Jr., Healey, J.M., Zheng, L., McCall, W., and Schulmeister, M.K., 1999, Hydrostratigraphic characterization of unconsolidated alluvium with direct-push sensor technology (abstract). GSA 1999 Annual Meeting Abstracts with Program. v. 31, no. 7, p. A350 (full report as Kansas Geol. Surv., Open-File Rept. 99-40) [available online]

Christy, C.D., Christy, T.M. and Wittig, V., 1994, A percussion probing tool for the direct sensing of soil conductivity; in, Proc. of the 8th National Outdoor Action Conf.: NGWA, p. 381-394.

Fetter, C.W., 2001, Applied Hydrogeology (4th Edition): Prentice Hall.

HydroSOLVE, Inc., 2001, AQTESOLV for Windows User's Guide.

Prescott, G.C., Jr., 1955, Geology and ground-water resources of Graham County, Kansas: Kansas Geological Survey, Bulletin 110. [available online]

Schmidt, G.W., and Marsi, K.L., 1961, Chemical analysis of ground water resources of Rooks County, Kansas: Kansas Academy of Science, Transactions 64, no. 1, p. 49-62

Springer, R.K., and Gelhar, L.W., 1991, Characterization of large-scale aquifer heterogeneity in glacial outwash by analysis of slug tests with oscillatory response, Cape Cod, MA: U.S. Geol. Surv., Water Res. Invest. Rep. 91-4034, p. 6-40.

Appendix: Well records

Links are to the Kansas Geological Survey's WWC5 database.

T-R-S Owner Well Depth Static Water Level Well Use Action Taken Completion Date Scan?
Sec. 7
NE SE NE
Kansas Dept. of Ag,
Div. of Water Resources
23.6 ft. 14.4 ft. Monitoring well/
observation/
piezometer
Constructed 16-Jul-2002 Scan
Sec. 7
NE SE NE
Kansas Dept. of Ag.,
Div. of Water Resources
58.5 ft. 14.5 ft. Monitoring well/
observation/
piezometer
Constructed 16-Jul-2002 Scan
Sec. 7
SE SE NE
Kansas Dept. of Ag.,
Div. of Water Resources
22.4 ft. 8.6 ft. Monitoring well/
observation/
piezometer
Constructed 16-Jul-2002 Scan

Kansas Geological Survey, Geohydrology
Placed online Aug. 30, 2007, original report dated Dec. 2002
Comments to webadmin@kgs.ku.edu
The URL for this page is http://www.kgs.ku.edu/Hydro/Publications/2002/OFR02_60/index.html